@aiclude/security-skill 0.1.0 → 1.0.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/SKILL.md +38 -25
  2. package/package.json +1 -1
package/SKILL.md CHANGED
@@ -1,57 +1,70 @@
1
1
  # /security-scan - AIclude Security Vulnerability Scanner
2
2
 
3
- AI Agent 시스템에서 사용하는 MCP Server 또는 Skill의 보안 취약점을 검사합니다.
4
- 기존 스캔 결과가 있으면 즉시 반환하고, 없으면 새로 등록하여 스캔을 수행합니다.
3
+ Scan MCP Servers and AI Agent Skills for security vulnerabilities. Returns existing scan results instantly if available, or registers the target and triggers a new scan automatically.
5
4
 
6
5
  ## Usage
7
6
 
8
7
  ```
9
- # 이름으로 보안 검색 (권장 - 기존 스캔 결과 활용)
8
+ # Search by name (recommended - leverages existing scan results)
10
9
  /security-scan --name <package-name> [options]
11
10
 
12
- # 로컬 경로 직접 스캔
11
+ # Scan a local directory directly
13
12
  /security-scan <target-path> [options]
14
13
  ```
15
14
 
16
15
  ## Parameters
17
16
 
18
- - `--name`: 검색할 MCP 서버 또는 Skill 이름 (npm 패키지명, GitHub 레포 )
19
- - `target-path`: 스캔할 로컬 경로 (--name 대신 사용)
20
- - `--type`: 대상 유형 (`mcp-server` | `skill`) - 자동 감지됨
21
- - `--profile`: 샌드박스 프로필 (`strict` | `standard` | `permissive`)
22
- - `--format`: 리포트 출력 형식 (`markdown` | `json`)
23
- - `--engines`: 사용할 스캔 엔진 (쉼표 구분)
17
+ - `--name`: Name of the MCP server or Skill to search (npm package name, GitHub repo, etc.)
18
+ - `target-path`: Local path to scan (use instead of --name)
19
+ - `--type`: Target type (`mcp-server` | `skill`) - auto-detected if omitted
20
+ - `--profile`: Sandbox isolation profile (`strict` | `standard` | `permissive`)
21
+ - `--format`: Report output format (`markdown` | `json`)
22
+ - `--engines`: Scan engines to use (comma-separated)
24
23
 
25
24
  ## Examples
26
25
 
27
26
  ```
28
- # MCP 서버 보안 검색
27
+ # Search security scan results for an MCP server
29
28
  /security-scan --name @anthropic/mcp-server-fetch
30
29
 
31
- # Skill 보안 검색
30
+ # Search scan results for a Skill
32
31
  /security-scan --name my-awesome-skill --type skill
33
32
 
34
- # 로컬 스캔
33
+ # Scan local source code
35
34
  /security-scan ./my-mcp-server
36
35
  ```
37
36
 
38
37
  ## What It Checks
39
38
 
40
- - **Prompt Injection**: LLM 대상 프롬프트 인젝션 패턴
41
- - **Tool Poisoning**: 도구 설명에 숨겨진 악성 지시문
42
- - **Command Injection**: exec/spawn 검증되지 않은 입력 사용
43
- - **Supply Chain**: 의존성의 알려진 CVE 악성 패키지
44
- - **Malware**: 백도어, 크립토마이너, 랜섬웨어 탐지
45
- - **Permission Abuse**: 과도한 파일시스템/네트워크 권한
39
+ - **Prompt Injection**: Hidden prompt injection patterns targeting LLMs
40
+ - **Tool Poisoning**: Malicious instructions embedded in tool descriptions
41
+ - **Command Injection**: Unvalidated input passed to exec/spawn calls
42
+ - **Supply Chain**: Known CVEs in dependencies and malicious packages (typosquatting)
43
+ - **Malware**: Backdoors, cryptominers, ransomware, data stealers, and obfuscated code
44
+ - **Permission Abuse**: Excessive filesystem, network, or process permissions
45
+
46
+ ## Scan Engines
47
+
48
+ 7 engines run in parallel for comprehensive coverage:
49
+
50
+ | Engine | Description |
51
+ |--------|-------------|
52
+ | SAST | Static code pattern matching against YAML rule sets |
53
+ | SCA | Dependency CVE lookup via OSV.dev, SBOM generation |
54
+ | Tool Analyzer | MCP tool definition analysis (poisoning, shadowing, rug-pull) |
55
+ | DAST | Parameter fuzzing (SQL/Command/XSS injection) |
56
+ | Permission Checker | Filesystem, network, and process permission analysis |
57
+ | Behavior Monitor | Runtime behavior pattern detection |
58
+ | Malware Detector | Signature scanning, entropy analysis, backdoor detection |
46
59
 
47
60
  ## Output
48
61
 
49
- 리포트에는 다음이 포함됩니다:
50
- 1. **위험 수준 요약** (CRITICAL / HIGH / MEDIUM / LOW / INFO)
51
- 2. **발견된 취약점 목록** (코드 위치, 설명, 심각도)
52
- 3. **사용 리스크 평가** (어떤 위험이 있는지)
53
- 4. **수정 권장사항** ( 취약점별 해결 방법)
62
+ The report includes:
63
+ 1. **Risk Level Summary** (CRITICAL / HIGH / MEDIUM / LOW / INFO)
64
+ 2. **Vulnerability List** (code location, description, severity)
65
+ 3. **Risk Assessment** (what risks are present and their impact)
66
+ 4. **Remediation Recommendations** (how to fix each vulnerability)
54
67
 
55
68
  ## Web Dashboard
56
69
 
57
- 모든 스캔 결과는 https://vs.aiclude.com 에서 확인할 수 있습니다.
70
+ View all scan results at https://vs.aiclude.com
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@aiclude/security-skill",
3
- "version": "0.1.0",
3
+ "version": "1.0.0",
4
4
  "description": "AIclude Security Vulnerability Scanner - Claude Code Skill for inline security scanning",
5
5
  "type": "module",
6
6
  "main": "./dist/index.js",